Large language models place language before genuine reasoning capabilities and can never achieve true understanding.